Measuring 24VAC current using hall effect or current transformer sensor

Thread Starter

DJ_AA

Joined Aug 6, 2021
206
Hi

I am working on a small project, where I need to be able to detect flowing current on the existing pump power control unit.

I would like to be able to detect when a pump has tripped(by the protection circuit) and pass this information remotely. At the moment, when every issue arise the LAMP is switched on and if I am correct I can simply measure the current flow


I would like to either use the standard clip-on of the current transformer sensor or the hall effect clip-on of the sensor.



As I am detecting the current going to a lamp @ 24VAC, would there be a minimum current requirement for sensors to accurately detect current flow? Or can it detect a few mA?


In addition to this, the pump is powered using 3 phases, can on clip-sensor be sufficient for that or would we need to use the sensor for each phase?
I would like to use a Non-Invasive Sensor as I do not want to play around in-depth with high power.

Is there any good manufacturer that I should be looking at? I found the following

https://www.poweruc.pl/collections/split-core-current-transformers2


Do these types of sensors require any calibration?

Are all the outputs generally analogue or can you get versions with maybe pulse or digital signal?

Thanks

This is the circuit that I need to add the sensors to.

IMG_5010.jpg



I need to measure High-Level Lamp, Overload Tripped Load lamp connection.
 

Irving

Joined Jan 30, 2016
3,190
I think the current in those lamps will be too small to reliably detect non-invasively. A typical 1W indicator will draw <45mA. The most sensitive current transformer SCT0006 rated 1A full scale will give 45uA output requiring some serious interfacing. You'll have the same issue with the ACS712 modules which are invasive; the most sensitive will give 8mV output for 45mA current.

You be better to add a resistor and the diode part of a Vishay AC optocoupler across each lamp to be monitored.
 
Last edited:

Thread Starter

DJ_AA

Joined Aug 6, 2021
206
Maybe I could create a PCB with hall effect IC on a PCB, which could be strapped around 24Vac wire, instead of the existing available sensor which works on much higher currents.

Non-invasively, is a priority, as I do not want to be removing wires or playing around disconnecting anything, i need to be able to connect the sensor very quickly, with no risk
 

GetDeviceInfo

Joined Jun 7, 2009
2,125
I worked on a process with saws. When the blades dulled, the motors would trip on overload. No problem as that was a PLC input. The problem was when blades were jammed, and a start was attempted, the O/C would trip. As there was no indication, the system could be started and things would go very bad. That was resolved by adding one of these (similar) to a single motor phase to confirm current at the motor.
That may or may not be a consideration with your application. Looking at your schematic, I would parallel a relay coil with each lamp that you are wanting to sense, with the contacts representing present condition. That won't account for an O/C trip, but that may not be an issue
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
206
If we were to add a parallel relay, then we could just add the following ACS712 modules, as we would then be able to detect the alerts, but I would prefer a Non-invasively option.
 

GetDeviceInfo

Joined Jun 7, 2009
2,125
If we were to add a parallel relay, then we could just add the following ACS712 modules, as we would then be able to detect the alerts, but I would prefer a Non-invasively option.
Do the OLxX, OLx, or MCx relays have spare poles? Often aux contacts can be hung on contactors.
With the ACS712 on the lamp circuit, all you are doing is confirming the lamp element. Something to consider when troubleshooting.

Wait, I'll take that back as you already have a parallel relay coil, ensuring a current path. You will have to resolve the ACS712 output.
 
Last edited:

Thread Starter

DJ_AA

Joined Aug 6, 2021
206
There are no poles, all I have at moment is this circuit diagram.

Yes, it confirms that lamp element, but that connection is only achieved once the float switch makes the connection. For example, if we want to detect the connection for the "High-Level Float", instead of removing the existing setup or indication, I am trying to have a parallel remote based system.

The priority is to detect:
High-Level Float
Overload Tripped Lamp's
 

GetDeviceInfo

Joined Jun 7, 2009
2,125
There are no poles, all I have at moment is this circuit diagram.

Yes, it confirms that lamp element, but that connection is only achieved once the float switch makes the connection. For example, if we want to detect the connection for the "High-Level Float", instead of removing the existing setup or indication, I am trying to have a parallel remote based system.

The priority is to detect:
High-Level Float
Overload Tripped Lamp's
There are no poles, all I have at moment is this circuit diagram.

Yes, it confirms that lamp element, but that connection is only achieved once the float switch makes the connection. For example, if we want to detect the connection for the "High-Level Float", instead of removing the existing setup or indication, I am trying to have a parallel remote based system.

The priority is to detect:
High-Level Float
Overload Tripped Lamp's
point being, a loss of lamp element = loss of signal. Just throwing out some ideas. Sounds like you have a path. :)
 

Reloadron

Joined Jan 15, 2015
7,081
My read on this so far is you have two large 3 phase pumps running on a 400 volt 50 Hz WYE supply.. Phase to phase is 400 volts and phase to neutral is the 240VAC for your control transformer which provides 24 VAC for all the control applications. Wire 1 is control hot and wire 0 is control neutral. With a focus on High Level float switch at a high level the switch closes passing voltage from Wire #1 to Wire #6, This will enable the high level red lamp. It will also enable an audible buzzer. Pressing the Cancel button will latch the cancel relay will silence the buzzer (no noise) but by design the red High Level Lamp will remain illuminated until the high level condition is relieved.

If you simply want to know when a high level state exist there are dozens of AC optocouplers out there which could parallel the existing lamp. This is one which comes to mind. Doesn't matter if the bulb fails. This is how I would likely go about it but there other ways. Trying to detect lamp current would work but it's a low current and if the lamp fails there will be zero lamp current.

With pump systems, especially large and very expensive systems I always incorporated 3 phase monitors at power entry. Lose a phase and I want shutdown.

Ron
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
206
My read on this so far is you have two large 3 phase pumps running on a 400 volt 50 Hz WYE supply.. Phase to phase is 400 volts and phase to neutral is the 240VAC for your control transformer which provides 24 VAC for all the control applications. Wire 1 is control hot and wire 0 is control neutral. With a focus on High Level float switch at a high level the switch closes passing voltage from Wire #1 to Wire #6, This will enable the high level red lamp. It will also enable an audible buzzer. Pressing the Cancel button will latch the cancel relay will silence the buzzer (no noise) but by design the red High Level Lamp will remain illuminated until the high level condition is relieved.

If you simply want to know when a high level state exist there are dozens of AC optocouplers out there which could parallel the existing lamp. This is one which comes to mind. Doesn't matter if the bulb fails. This is how I would likely go about it but there other ways. Trying to detect lamp current would work but it's a low current and if the lamp fails there will be zero lamp current.

With pump systems, especially large and very expensive systems I always incorporated 3 phase monitors at power entry. Lose a phase and I want shutdown.

Ron
That is correct. I will be using a parallel AC optocoupler. I am not sure, should I use available module available in the market, or actually add it to my PCB. The advantage of the available modules will be it is easily replaceable and have no AC on the PCB, but would I guess it would be mean some extra wiring in the enclosure.

I would also like to monitor the 3 phase connection, and maybe work out the power consumption or also detect any pump issues from the data pattens. For this, I believe I will a non-invasive sensor would work.

So, would a standard would a AC current sensor, SCT013-10A-1V / # F T0R 0775 | eBay be sufficient or would I need to use a hall effect sensor?

So i simple strap this around all 3 wires or each wire has its own sensor?
 

Reloadron

Joined Jan 15, 2015
7,081
Starting with power to your pumps and turn key solutions. If you just want to know if each pump is seeing all three phases I have used these non invasive sensors. I have used them on heating elements and pumps. Moving up from there there are turn key solutions like these units. The latter example allows a user to have pre-set levels for detecting things like phase reversal, phase loss, phase unbalance, overvoltage and undervoltage protection. The links are merely for examples as things like this are available from many manufacturers with a wide cost range. For accurate measure of three phase current there are three phase AC current transducers. These will output a low voltage of current proportional to the measured current. All of these are pretty much off the shelf turn key solutions. Depending on budget they range from inexpensive to expensive.

Roll your own solutions. Each line will need it's own sensor and the current transformer you linked to, (SCT013-010 10A 1V Non-invasive Split Core Current Transformer Sensor) should do fine. Using a split core avoids having to cut any wires. These are very common these days. There is a caveat. The primary is of course AC. Most have a built in burden resistor. So for example 10.0 Amps AC affords 1.0 volt AC out. Normally one needs some conditioning at this point to get from what we have to what we want. The 1.0 VAC out will be a peak to peak wave varying above and below a 0.0 volt reference. You need to decide what you wish to do with that signal out. So they do work but the output needs some signal conditioning.

Verse using a current transformer will a hall effect sensor work? Yes it will. Pretty sure I saw mention of the ACS 712 or ACS 723 manufactured by Allegro. Many are available on breakout boards making for easier implementation. Again the hall effect sensor will have an output which follows the input so AC in and AC out requiring some degree of signal conditioning.
Here is an ACS 712 data sheet. Just click the data sheet button on the linked page. They come in +/- 5, 10 and 30 amp versions. Allegro I think also has some newer versions out and sensors like this are available from several sources already mounted on a board in modular form.

So overall there are a number of ways to get where you want to be, eash having considerations. During my years I liked turn key solutions and reliability because if something did fail we wanted a solution on the shelf with minimal down time. We also had restrictions on what we could buy.

Ron
 

Thread Starter

DJ_AA

Joined Aug 6, 2021
206
Thanks, the primary aim of the project is remotely detecting the alerts, but as a bonus, I am looking into the possibility of reading the 3 phase power lines. This done need to be extremely accurate, but something that can be fairly accurate to give information such as power usage and see certain patterns of any issue that could arise.

I am looking for a split-core solution as I do not want to be tempering with live 3 code wires.:(

These sensors are available in multiple current ratings, for example, 5A or 10A if for example, the pump is consuming about 3.7A, then I guess it would be more viable to use the 5A so that I can maximise the 1V range in my ADC. So when you mention signal conditioning, can you explain a bit further? Is it not as simple as connecting it directly to an ADC pin on uC and if possible keeping the reference to a value slightly above 1V?

In regards to the clamp style current sensors, is there a particular make that is recommended to us, for example, I saw that TDK also has their own version and also a company called Seeed Technology Co., Ltd that is slightly lower in price.
 

Reloadron

Joined Jan 15, 2015
7,081
I am looking for a split-core solution as I do not want to be tempering with live 3 code wires.:(

These sensors are available in multiple current ratings, for example, 5A or 10A if for example, the pump is consuming about 3.7A, then I guess it would be more viable to use the 5A so that I can maximise the 1V range in my ADC. So when you mention signal conditioning, can you explain a bit further? Is it not as simple as connecting it directly to an ADC pin on uC and if possible keeping the reference to a value slightly above 1V?
OK, here is what you are looking at. Little clamp on current transducers are fine but when I mention signal conditioning here is what I am getting at. Your pumps are three phase AC. A current transformer needs a load, this is called a burden resistance to generate a signal across. Your motor supply voltage is 3 phase 400 volt 50 Hz. Typical motor current is 3.7 amps on each phase. The overloads look to be adjustable 3.7 to 5.5 amps on each phase. The voltage and current are the RMS values. You have a sine wave which goes above and below a 0 volt level. OK, so we add a current transformer clamp on to each phase, the transformers are 5 Amp = 1 Volt using an internal burden resistance. The waveform out is still an AC waveform and still changing above and below 0 volts. Actually you have three of them 120 degrees apart. Consider also 0 to 5 amps or 0 to 10 amps RMS AC depending on CT selection is now 0 to 1.0 VAC RMS. So you need to scale things.

So we run that signal into a uC (micro-controller) but unless the uC can take a differential analog input you need to offset the input to the uC so the portions of the sine below zero volts can be passed along. There are basic circuits to do this. Just remember more parts and more design. If for example the ADC has a 0 to 5 volt input you want to offset by about 2.5 VDC. Next you write the code to read the RMS value of your sine wave input. Not difficult but not easy either. This is the world of signal conditioning, what you have to what you want. Been years since I did this stuff but another potential problem each CT output can't share the same common so your ADC (analog to digital converter) will need to have a differential input for each phase since you motors start with Delta configured 3 phase. The voltages and currents are phase 1 to phase 2, phase 2 to phase 3 and phase 3 to phase 1. Your voltages and currents are phase to phase. Now if I have this wrong another member please do point this out but thinking about it I am pretty sure you can't have a single common.

Next and it won't be cheap or inexpensive is to go with the turn key modules I mentioned and if you want voltage they make transducers for that. Modules like this afford galvanic isolation and are accurate.
Is it not as simple as connecting it directly to an ADC pin on uC and if possible keeping the reference to a value slightly above 1V?
Unfortunately no, not that simple. You can view the data in real time but if you want to save any for review at a later time you get into data logging and additional cost. Collecting data is like collecting anything else. You need a place to put it. :(

Ron
 
Top